I am working on a thesis using a driving simulation experiment. The sample needs to represent the population accurately. However, when dealing with a small sample size, such as 30 to 40 individuals, how should we account for factors like age, gender, and other variations during the sampling process? Is it necessary to assume that the sample is inherently biased, and should the subsequent analysis consider that? Any assistance would be greatly appreciated.
Thank you, Wilhelm, for your insightful responses. I am conducting a driving simulation experiment, and my target population includes any individual with a valid driver’s license, both local residents and international visitors residing temporarily in the city where the laboratory is located.
Given the effectively unbounded nature of this population and the absence of a practical sampling frame, I am struggling with the methodological question of how to approximate random sampling under these real-world constraints.
Specifically, how can one select participants in a manner that meaningfully approximates randomness while accounting for potential self-selection bias inherent in convenience sampling?
Are there defensible strategies for balancing feasibility with statistical rigor when the population is heterogeneous, geographically dispersed, and partially transient?
Thank you, Wilhelm. I will take your suggestion into account. The reason for including international visitors is that the city has a notable number of temporary international residents and students.
- Badges
- Science topic
- Similar topics
- Psychology